OpenAI unveils GPT-5.6-Cyber, a model built for cybersecurity
Daybreak program splits into Blue and Red tiers, with access restricted to vetted defenders
00
Tag
security
Anthropic mandates watermarking on Claude outputs worldwide
Under EU AI Act rules, invisible markers for text and signed provenance data for images start August 2026
1170
Apple unveils technique to block fine-tuning of AI model weights
"DLR-Lock" research lets users run open-weight models as-is while blocking modification
10
Cohere CEO: "Open-source models need three things"
Aidan Gomez explains why the company is expanding its open-source lineup and how it's designing for "sovereign" deployment
10
Atlassian AI Agent Breached via White Text Hidden in PDF
Security firm PromptArmor discloses vulnerability letting Rovo exfiltrate Jira and Confluence data without a trace
00
Claude-based AI assistant breaches gym system to jump waitlist
An Australian developer's OpenClaw agent found an authentication flaw and manipulated a waitlist ranking
20
Startup Launches 'Jamming Device' Amid AI Wearable Eavesdropping Concerns
As AI recording devices in glasses and necklace form spread, one company has developed a jamming device in response
10
US community colleges see rise in AI-driven fake student financial aid fraud
According to a New Yorker report, fraudsters are enrolling fake students, using AI to complete their coursework, and collecting financial aid payments
00
That's the last story.